Expasion due to Heat

Elmer cases by the users for the users
kevinarden
Posts: 2316
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Expasion due to Heat

Post by kevinarden »

NAFEMS test case, hot spot.
problem.PNG
(91.75 KiB) Not downloaded yet
NAFEMS answer is 50 MPa

abaqus tria6 54.6
Elmer tria6 54.5
abaqus quad8 51.5
Elmer Quad8 49.7
expansion.zip
(145.39 KiB) Downloaded 65 times
stress.PNG
(170.19 KiB) Not downloaded yet
Rich_B
Posts: 423
Joined: 24 Aug 2009, 20:18

Re: Expasion due to Heat

Post by Rich_B »

Hello Kevin,

I ran the two tests, and triangles ran fine, but quads did not. Maybe you could take a look?

I found this example of what looks like the same case:

https://help.altair.com/hwsolvers/os/to ... m_t1_r.htm

Thanks, Rich.

Edit: Windows 10, Elmer binaries installer dated 17 October 2023.
kevinarden
Posts: 2316
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Expasion due to Heat

Post by kevinarden »

It is the same case although they do symmetry planes.
I ran the quad case again,
The quad case runs fine for me. W10 V9.0 Compiled 9/6/23

49.7 MPa
Rich_B
Posts: 423
Joined: 24 Aug 2009, 20:18

Re: Expasion due to Heat

Post by Rich_B »

Hi,

I just downloaded again the expansion.zip archive from the first post. Extract the archive to a folder, and there are two elmer mesh folders, trias and quad, and one sif file, mcase.sif. Without editing anything, enter 'elmersolver mcase.sif'. Here's the error message:
ERROR:: IterSolve: Numerical Error: System diverged over maximum tolerance.
Note: The following floating-point exceptions are signalling: IEEE_INVALID_FLAG IEEE_DIVIDE_BY_ZERO
STOP 1
Next, open the quad mesh folder in ElmerGUI. The attached screenshot shows body 1 is the outer circular portion, which is set with body force 1 to 100. Maybe the wrong quad mesh folder was included in the archive?

Rich.
Attachments
quads.png
quads.png (192.46 KiB) Viewed 949 times
kevinarden
Posts: 2316
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Expasion due to Heat

Post by kevinarden »

When I download the zip Body 1 targets mesh body 2

Body 1
Target Bodies(1) = 2
Name = "Body 1"
Equation = 1
Material = 1
Body Force = 1
Initial condition = 1
End
kevinarden
Posts: 2316
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Expasion due to Heat

Post by kevinarden »

I downloaded the posted zip, it is set to the quad mesh, ran it with no changes, it ran fine.
Rich_B
Posts: 423
Joined: 24 Aug 2009, 20:18

Re: Expasion due to Heat

Post by Rich_B »

I did a complete uninstall of Elmer, and then installed Elmer binaries using the Windows installer dated 14 Sept 2023. Same error message occurs. Very bizarre.

Rich.
kevinarden
Posts: 2316
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Expasion due to Heat

Post by kevinarden »

I just tried it on Ubuntu and it works, perhaps post the whole log file, maybe increase the message level above 5
Max Output Level = 5
Rich_B
Posts: 423
Joined: 24 Aug 2009, 20:18

Re: Expasion due to Heat

Post by Rich_B »

Okay, attached is an archive with the files and the solver log, output = 50.

Thanks, Rich.
Attachments
expansion2.zip
(44.27 KiB) Downloaded 38 times
kevinarden
Posts: 2316
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Expasion due to Heat

Post by kevinarden »

I downloaded your zip and ran it as is it ran fine. Confusing?
Post Reply